Awareness, knowledge and experiences of women regarding cervical cancer in rural KwaZulu-Natal, South Africa

(1)

AWARENESS, KNOWLEDGE AND EXPERIENCES OF

WOMEN REGARDING CERVICAL CANCER IN RURAL

KWAZULU-NATAL, SOUTH AFRICA

Beauty Hlengiwe Ndlovu

Thesis presented in partial fulfilment of the requirements for the degree

of Master of Nursing Science in the Faculty of Health Sciences at the

Stellenbosch University

Supervisor: Dr ELD Boshoff

(2)

DECLARATION

By submitting this thesis electronically, I declare that the entirety of the work contained therein is my own, original work, that I am the sole author thereof (save to the extent explicitly otherwise stated), that reproduction and publication thereof by Stellenbosch University will not infringe any third party rights and that I have not previously in its entirety or in part submitted it for obtaining any qualification.

23 March 2011

(3)

ABSTRACT

Title

Awareness, knowledge and experiences of women regarding cervical cancer in rural Kwa-Zulu-Natal, South Africa.

Background

Cervical cancer has been identified as the second most common cancer in women and contributes to the high mortality rate in women. Among all cancers in women, cervical cancer is rated the second most common cancer in women worldwide. In poorly resourced settings, access to services offering cervical screening is still a challenge and it is estimated that more than 50% of women in developing countries have never had a single screening test for cervical abnormalities.

Purpose

The purpose of this study was to assess women’s awareness, attitudes and experiences regarding cervical smear testing and for cervical cancer in rural KwaZulu-Natal and to better understand factors influencing access to and utilization of cervical cancer screening services by rural women.

Methods

The method employed was a descriptive study using a questionnaire to collect quantitative data. The sample consisted of 69 women aged 30 years and above, was taken from women who were enrolled in the on-going Microbicide Clinical Trial and attending follow-up clinic visits between July and August 2009. The primary outcome measure for the analyses was who has been screened for cervical cancer and this was assessed from the previous history reports of the women. The secondary outcome measure was to investigate knowledge and perceptions regarding cervical cancer and screening. Socio-demographic factors associated with having been screened were also explored.

(4)

Results

Out of 69 women, only N=13 (18.8%) reported ever screening for cervical cancer. More than half of women who had never screened reported lack of information as a barrier to screening N=50 (71.4%). Older women aged 35-45, 45 and above were less likely to screen compared to women aged 30 to 34 years of age (OR: 0.06). Having an educational background seemed to increase the likelihood to screen, twice if a woman had primary education (OR 2.0) and almost three times (OR 2.67) if a woman had a secondary or a higher education. More than half of the respondents considered themselves at risk for cervical cancer N=42 (60.8%) and almost all showed a willingness to screen in the future N=64 (93%).

Conclusion

Most of the women in this study had never been screened for cervical cancer in their lifetime as reflected by n=55 (82%) while only n=14 (18%) ever screened for cervical cancer. The results of this study cannot be generalised to the population due to the small sample size. However, there is need to facilitate comprehensive health education and the implementation of cervical screening programmes to target women in rural communities to contribute to the success of the cervical screening programme. The results of this study showed that 60% of respondents were informed by health care professionals on cervical cancer screening. Health care workers also should play a vital role in educating communities on cervical cancer and on the benefits for cervical cancer screening, through reaching all patients who utilise health care services with cervical cancer information and also communities through outreach programmes.

(5)

OPSOMMING

Titel: Vrouens se bewustheid, houding en ervarings van smeertoetse en servikale karsinoom in die landelike gebiede van KwaZulu-Natal

Agtergrond

Servikale kanker is geïdentifiseer as die tweede mees algemene karsinoom in vrouens en dra by tot die hoë sterftesyfer in vrouens. Van al die tipes karsinoom wat by vrouens voorkom, is servikale karsinoom die tweede mees algemene karsinoom onder vrouens wêreldwyd. Die beskikbaarheid van dienste wat servikale smeer toetsing bied, is nog steeds ’n uitdaging in arm gebiede en daar word geskat dat meer as 50 % van vrouens in ontwikkelende lande nog nooit ’n toets vir enige servikale abnormaliteite gehad het nie.

Doel

Die doel van hierdie studie was om vrouens se bewustheid, houding en ervarings van servikale smeer toetsing en van servikale karsinoom in die plattelandse gebiede van KwaZulu-Natal te toets en om ’n beter begrip te kry van faktore wat ’n invloed het op toegang tot en gebruik van servikalesmeer toetsing by vrouens in landelike areas.

Metode

Die metode wat gebruik is, is ’n beskrywende studie waarin gebruik gemaak is van vraelyste om kwantitatiewe data te versamel. Die monster het bestaand uit 69 vrouens, ouderdom 30 jaar en ouer, wat deelnemers was aan die “Microbicide Kliniese Navorsingsprojek” en wat opvolgbesoeke by klinieke gehad het tussen Julie en Augustus 2008. Die primêre bevinding, wie al ooit vir servikale karsinoom getoets is, is bereik deur die inligting in die laboratorium verslae van die vroue na te gaan. Die sekondêre bevinding was om die deelnemers se kennis en persepsies aangaande servikale karsinoom te toets. Sosio-demografiese faktore wat verband hou met of deelnemers ooit getoets is, is ook ondersoek.

(6)

Resultate

Van die 69 vrouens, het slegs N=13 (18.8 %) gerapporteer dat hulle ooit getoets is vir servikale kasinoom. Meer as die helfte van die vrouens wat ooit getoets is vir servikale karsinoom het gerapporteer dat ’n gebrek aan inligting ’n weerhoudende faktor was tot die toetse, N=50 (71.4%). Ouer vrouens tussen die ouderdom van 35 – 45, 45 en ouer was minder bereid om te toets in vergelyking met vrouens tussen die ouderdom van 30 tot 34 (OR: 0.06). Dit blyk asof skoolonderrig die kanse op toetsing verhoog, vrouens met primêre skoolopleiding se kanse dat hulle getoets is, is twee keer groter (OR 2.0) en amper drie keer meer (OR 2.67) as ’n vrou sekondêre onderrig of hoër onderrig ontvang het. Meer as die helfte van die respondente dink hulle loop ’n risiko om servikale kanker te kry N=42 (60.8%) en feitlik almal was bereid om hulle te laat toets in die toekoms N=64 (93 %).

Bevinding

Die meeste vroue in hierdie studie n=55 (82%) was nog nooit in hul leeftyd getoets vir servikale karsinoom nie terwyl slegs n=14 (18%) ooit getoets was vir servikale karsinoom. Die resultate van hierdie studie kan nie veralgemeen word nie, aangesien die navorsings-populasie as gevolg van die klein steekproef te klein was. Nietemin is daar ‘n behoefte vir die fasilitering van omvattende gesondheidsopvoeding en die implementering van servikalesmeer toetsing programme. Die resultate van hierdie studie het aangedui dat 60% van die respondente deur professionele gesondheids werkers ingelig is met betrekking tot servikalesmeer toetsing. Gesondheidswerkers behoort ‘n vitale rol te speel in die opvoeding van gemeenskappe in verband met servikale karsinoom en die voordele van hiervan deur alle pasi nte wat gesondheidsdienste benut in te lig omtrent servikale karsinoom en ook deur middel van gemeenskaps-uitreikings programme.

(7)

ACKNOWLEDGEMENTS

I would like to extend my appreciation to the following people who have contributed to the completion of this study:

 Dr E.L.D. Boshoff, my supervisor, for her guidance and support;

 The Microbicide Clinical Trial participants who agreed to take part in this study;  Mitzy Gafos, for the (STATA) statistical programme training;

 The University of Stellenbosch Statistician, Dr Kidd, for statistical support;  Professor C.T Moyo, for proof-reading the chapters;

 The Microbicide Development Programme colleagues, for their motivation and support; and

(8)

ACRONYMS AND ABBREVIATIONS

CI Confidence Interval

Pap Papanicolau smear test HPV Human Papilloma Virus

OR Odds Ratio

(9)

LIST OF TABLES

Table 4.1: Age-Group in women enrolled in cervical cancer study in rural KwaZulu-Natal n=69 ... 26 Table 4.2: Educational Status in women enrolled in cervical cancer study in rural

KwaZulu-Natal n=69 ... 26 Table 4.3: Employment status in women enrolled in cervical cancer study in rural

KwaZulu-Natal n=69 ... 27 Table 4.4: Monthly Earnings in women enrolled in cervical cancer study in rural

KwaZulu-Natal n=18 ... 27 Table 4.5: Sexual debut age in women enrolled in cervical cancer study in rural

KwaZulu-Natal n=67 ... 27 Table 4.6: Marital status in women enrolled in cervical cancer study in rural

KwaZulu-Natal n=69 ... 28 Table 4.7: Previous pregnancies in women enrolled in the cervical cancer study in rural

KwaZulu-Natal n=69 ... 28 Table 4.8: Contraception in women enrolled in the cervical cancer study in rural

KwaZulu-Natal n=49 ... 29 Table 4.9: The report of previous screening for cervical cancer characterized per

socio-demographic characteristics in rural women (69) aged 30 years and above enrolled in the study. ... 29 Table 4.10: Factors associated with having been screened for cervical cancer in women

aged 30 years and above in rural KwaZulu-Natal ... 31 Table 4.11: Cervical cancer description by women aged 30 years and above enrolled in the

study survey in rural KwaZulu-Natal n=69 ... 33 Table 4.12: Knowledge of cervical cancer risk factors, screening and prevention by

women aged 30 years and above, enrolled in the cervical cancer survey in rural

KwaZulu-Natal n=69 ... 34 Table 4.13: Cervical cancer screening perceptions and women’s knowledge thereof

(13)

LIST OF FIGURES

Figure 4.1: The main source of cervical cancer information reported by women enrolled in the cervical cancer study in rural KZN. ... 34

(14)

CHAPTER 1:

ORIENTATION TO THE STUDY

1.1 INTRODUCTION

Cervical cancer forms in the tissues of the cervix (the organ connecting the uterus and vagina). It is usually a slow-growing cancer that may not have symptoms but can be found with regular Pap tests (a procedure in which cells are scraped from the cervix and looked at under a microscope). Cervical cancer is almost always caused by Human Papillomavirus (HPV) infection. Cervical cancer is malignant neoplasm of the cervix uteri or cervical area. It may present with vaginal bleeding but symptoms may be absent until the cancer is in its advanced stages (WHO/ICO Information Centre on HPV and Cervical Cancer, 2009). Treatment consists of surgery (including local excision) in early stages, and chemotherapy and radiotherapy in advanced stages of the disease.

Pap smear screening can identify potentially precancerous changes. Treatment of high-grade changes can prevent the development of cancer. In developed countries, the widespread use of cervical screening programs has reduced the incidence of invasive cervical cancer by 50% or more (American Cancer Society, 2009:10). Human Papilloma Virus (HPV) infection is a necessary factor in the development of nearly all cases of cervical cancer.

Cervical cancer is a typically slow-growing cancer that may not have symptoms but can be early diagnosed by doing a Pap smear. The burden of disease due to transmissible diseases such as HIV and Human Papilloma Virus (HPV) is increasing especially in developing countries. Research has shown that HPV is the cause of about 70% of all cervical cancers (WHO/ICO, 2010). Leroy, Ladner, De Clercq, Meheus, Nyiraziraje, Karita and Dabis (1999:104), and Tate and Anderson (2002:881) have shown a statistical significant relationship between HIV, HPV and cervical abnormalities. According to the World Health Organization (WHO) Information Centre on HPV and Cervical Cancer, 2009 report, cervical cancer is currently the primary cancer in women in South Africa with an annual new case incidence of 6 742, in African women. It is estimated that the annual number of new cases worldwide, is as high as 493 243. Deaths due to cervical cancer amount to 273 505 women worldwide and 3 681 in South Africa (WHO/ICO Information Centre on HPV and Cervical Cancer, 2009).

(15)

Knowledge about the disease and risk factors are therefore important in determining appropriate health seeking behaviours with the aim to prevent invasive cancer and reduce mortality rates. Chapter one orientates the reader to the rationale of the problem and an overview of the research methodology on familiarity of women regarding cervical cancer. This research study will provide insight into the knowledge level and experiences of rural women to screening and cancer, and also explore the factors influencing access to screening services.

1.2 RATIONALE

The South African policy on cervical screening recommends that all women over 30 years of age should have three free Pap smears at ten-year intervals. This screening campaign is aimed to achieve 70% coverage by 2010 (Kawonga & Fonn, 2008:32). It is regrettable to state that coverage remains sub-optimal and it is estimated that more than 50% of women in South Africa never had a single screening test for cervical abnormalities.

A study conducted in rural KwaZulu-Natal to determine baseline data regarding knowledge about risk factors for cervical cancer showed a low uptake of the Pap smear test and a low level of knowledge on the prevention of cervical cancer, risk factors, and the uptake of cervical screening. The study reported that only 6% had knowledge on all the risk factors while 65% knew only one of the risk factors of cervical cancer. Less than half (49%) of these participants correctly stated that the Pap smear screening test is used for early detection of cervical cancer and only 18% had ever been screened for cervical cancer (Hoque, Hoque & Kader, 2008:112).

A further concern is mentioned by Bärnighausen, Tanser & Newell (2009:407), where they state that there is no decline in the high HIV incidence in Kwa-Zulu Natal, which strongly relates to the higher incidence of HPV and cervical cancer in the area.

1.3 PROBLEM STATEMENT

The problem identified is that more than 50% of rural women in South Africa have not been exposed to cervical screening and there is a lack of knowledge regarding risk factors related to cervical cancer and the screening for and prevention of cervical cancer.

(16)

1.4 RESEARCH QUESTION

The question that the researcher wishes to answer is what are the factors influencing women’s awareness, knowledge and experience of cervical cancer and cervical cancer smear screening in the rural population of KwaZulu-Natal?

1.5 AIM OF THE STUDY

The overall goal of the study was to explore and describe factors contributing to women’s awareness, knowledge and experiences regarding cervical cancer and cervical cancer smear screening.

1.6 OBJECTIVES

The specific objectives were to

 determine the rate of screening for cervical cancer among women aged 30 years and above in rural KwaZulu-Natal,

 explore and describe the knowledge and perceptions of women on cervical cancer and cervical cancer smear screening in this rural population.

 identify and characterise socio-demographic factors associated with accessing cervical cancer screening services by women aged 30 years and above,

1.7 RESEARCH METHODOLOGY

The methodology used in this study was a descriptive design with a quantitative approach. The self-designed structured questionnaire was used to collect quantitative data on awareness, knowledge and experiences of rural KwaZulu-Natal women on cervical cancer and cervical cancer screening. The sample was a convenience sample as eligible women, aged 30 years and above, who were already randomized to the Microbicide clinical trial at the three UMkhanyakude district clinics were asked to participate as they came to the clinics.

The research instrument was piloted on a sample of five women who were comparable to sample respondents in terms of socio-demographic and geographic characteristics. This was conducted in July 2008 at Somkhele Clinic.

1.8 DEFINITIONS

(17)

Awareness

Awareness is described as appreciation, familiarity, knowledge, observation or understanding (Oxford Concise English Dictionary 1995). For this study awareness meant “being familiar and also knowledgeable about cervical cancer and cervical cancer smear screening.” It also relates to the experience and perceptions influencing the uptake of cervical screening services.

Cervix

The cervix is the lower part or neck of the uterus forming the opening to the vagina. It is divided into 2 parts, namely the endo-cervix, internal part and ecto-cervix, the outer part that is next to the vagina (Pocket Medical Dictionary, 2003: 57).

Cervical cancer

Cervical cancer relates to the actual neoplasma cancerous cell changes in the cervix commonly referred to as carcinoma in situ (cancerous growth localised) and invasive cancer (cancer spreads to nearby organs).

Cervical screening

For the purpose of this study cervical screening relates to early detection of pre-cancer lesions through a Papanicolau smear (Pap).

A Papanicolau test is a screening tool used to detect cervical abnormalities. Mucus and cells are collected from the ectocervix and endocervix, by scraping and then fixed onto a glass slide and sent to the Cytopathology laboratory for assessment (American Cancer Society, 2009: 10).

1.9 SUMMARY

Chapter one has given the introduction and orientation of the study, the research problem, goals, methodology and design, the ethical considerations and the further outlay of the research report, as well as the operational definitions. The next chapter will discuss a comprehensive review of literature on the topic and related matters of the study.

(18)

CHAPTER 2:

LITERATURE REVIEW

Chapter one addressed the orientation to the study as well as the purpose thereof, and a synoptic overview of the research methodology that was implemented in this research. A detailed literature search was conducted using Pub Med, and the relevant literature was reviewed. The output of the review is aimed at conceptualising the history and present status of cervical cancer, the uptake of screening services and factors influencing access. Where relevant, studies conducted in other countries were reviewed and important findings briefly summarised to explain the extent of the problem and related factors in other areas. This chapter provides a comprehensive overview of the literature on the topic: Awareness, Knowledge and Experiences of women regarding Cervical Cancer.

2.2 CERVICAL CANCER

Cancers can be caused by DNA (Deoxyribo-Nucleic Acids) mutations (gene defects) that activate cells promoting cell division (oncogenes). Sometimes this could be caused by inactivation of tumour suppressor genes, resulting to abnormal proliferation of cervical cells. Cancer of the cervix occurs as a result of abnormal cell changes in the tissue layers of the cervix. It may appear as a cauliflower-like growth that bleeds easily on contact (Smeltzer & Bare, 2004:1379).

Hawes, Critchlow, Faye Niang, Diouf, Aissatou, Touré, Kasse, Dembele, Coll-Seck, Kuypers and Kiviat, (2002:555) and Khan, Castle, Lorincz, Wacholder, Sherman, Scott, Rush, Glass and Schiffman, (2005:1073) have shown that the most common risk factor for cervical cancer is the Human Papilloma Virus (HPV), which is thought to produce proteins that cause proliferation of cervical lining cells leading to cancer (American Cancer Society, 2009:1). Cervical cancer usually starts in the cells on the surface of the cervix, involving columnar or squamous cells. The cells mostly affected are squamous cells. Progression of pre-cancer lesions (dysplasia) is usually slow and detection of early pre-cancer lesions can easily be achieved through Pap smear testing and is 100% treatable (Cervical Cancer Health Guide, 2008:2). If left undetected, pre-cancer lesions can develop into cervical cancer and eventually

(19)

spread to adjacent organs such as the uterus and ovaries, bladder, intestines and liver. The Human Papilloma Virus (HPV) spreads through sexual intercourse and is responsible for almost all cervical cancers (Cervical Cancer Health Guide, 2008:2). In countries where resources are allocated to prevention initiatives the prevalence and mortality of cervical cancer is lower, for an example in developed countries. Screening for cervical cancer is important so as to ensure early detection of pre-cancer lesions to reduce the incidence of mortality due to cervical cancer in the population (Adanu, 2002: 286). Knowledge and the attitude of the women to smear testing are important in improving the understanding of factors influencing the uptake of diagnostic services available.

2.2.1 Cervical cancer risk factors

Different cancers have various risk factors. It is important to note that although the presence of cervical cancer risk factors increase the odds of a woman getting cervical cancer in her lifetime, not all women exposed to risk factors develop cancer. Several risk or predisposing factors were identified and discussed.

2.2.1.1 Human Papilloma Virus (HPV)

The most important risk factor for cervical cancer is the sexually transmitted HPV. HPV infection is mostly prevalent in young women under 30 and less in older women over 30 years. There is currently no cure for HPV infection apart from management of lesions or growths caused by HPV infection (American Cancer Society, 2009:1). HPV type 16 and type 18 cause approximately 70% of incidences of cervical cancer worldwide, although there are variations across countries. Early studies have shown a relationship between HPV and cervical neoplasma. In a study conducted in Denmark 1998, women with high risk of HPV infection had a 33-fold increased risk of cervical neoplasma compared to HPV-negative women. In the same study it was found that HPV test has a higher sensitivity than a Pap smear test (Kjaer, 1998:2). Women who were HPV positive for type 16 and 18 were found to be at a higher risk of developing cervical cancer lesions when followed over a 10-year period (Khan, Castle, Lorincz, Wacholder, Sherman, Scott, Rush, Glass and Schiffman, 2005:1073). This further suggests evidence of HPV as a significant risk factor for cervical cancer.

2.2.1.2 Weakened Immune System

Chirenje, Loeb, Mwale, Nyamapfeni, Kamba and Padian (2002:766); Hawes, Critchlow, Faye Niang, Diouf, Aissatou, Touré, Kasse, Dembele, Sow, Coll-Seck and Kiviat (2003:566); and Leroy, Ladner, De Clercq, Meheus, Nyiraziraje, Karita and Dabis (1999:104), concur that the

(20)

highest burden of pre-cancer lesions is among HIV infected women and in some cases co-existence of HPV infection also increases this risk. In addition to HIV seropositivity as a risk factor, Ruche, Ramon, Mensah-Ado, Bergeron, Diamondé, Sylla-Koko, Ehouman, Touré-Coulibaly, Welffens-Ekra and Dabis (1998:2403), argue that other factors significantly associated with cervical lesions should also be considered, such as smoking, parity (given birth) three times or more three, low educational level and non-usage of oral contraceptives. Conversely, early sexual debut and multiple sexual partners, which are known to be risk factors for cervical cancer, were not associated with increasing risk in this study (Ruche et al., 1998:2403). Some studies have shown a significant association between recurrences of cervical dysplasia after therapy or a hysterectomy in women who are HIV infected (Tate & Anderson, 2002:881). Therefore, there is an increasing concern for cervical cancer screening for women living with HIV, though this remains a challenge for low-resource settings. The National Health Departments should encourage Pap smear testing for all patients who are HIV infected to facilitate early diagnosis and prompt management.

2.2.1.3 Smoking

The risk of developing cervical cancer is doubled in smoking women compared to non-smokers due to exposure to high levels of carcinogens in smoke. Carcinogens are thought to damage the DNA of cervical cells, leading to cervical cancer (American Cancer Society, 2009:2). It is of interest to learn how women smokers and non-smokers perceive their risk. In a study conducted among 722 women ranging between 20 and 64 years, smokers did not perceive themselves to be at high risk. The intention to attend a cervical screening was similar among both groups of smokers and non-smokers (Marteau, Hankins and Collins, 2002:20); although smoking among African women is not highly prevalent, there is a need to raise awareness in all communities in this regard. In South Africa smoking among women contributes 7.70% of cervical cancer (WHO/ICO Information Centre on HPV and Cervical Cancer, 2007:1).

2.2.1.4 Other risk factors

The other risk factors include: a family history of cervical cancer, early sexual debut, age, a lack of regular Pap smear testing, as well as the number of pregnancies (American Cancer Society, 2009:4; Ruche, Ramon, Mensah-Ado, Bergeron, Diamondé, Sylla-Koko, Ehouman, Touré-Coulibaly, Welffens and Dabis, 1998:2401). In a study conducted in Mali among women with invasive cancer, an increase in the number of pregnancies increased the risk of

(21)

cervical cancer, this relationship was statistically significant. In addition early sexual debut increased odds of invasive cancer (Bayo, Bosch, de Sanjose, Munoz, Combita, Coursaget, Diaz, Dolo, van den Brule and Meijer, 2002:206).

2.2.1.5 The link between cervical cancer and HIV and AIDS

The HIV incidence remains high in this rural KwaZulu-Natal population and has remained high over a 5 year period, from 2003-2007. The results from a large population-based longitudinal surveillance in this rural population showed the overall incidence of 3.4 per 100 person-years (Confidence interval 3.1-3.7). These results were similar to those reported in a longitudinal population-based study in another rural population, 4.9 and 2.2 per 100 person-years in women and men respectively (Bärnighausen, Tanser & Newell 2009:407). Both HIV and cervical cancer are regarded as sexually transmitted infections which initially do not show symptoms until the progression of the disease. Research studies suggest that the lifespan of women living with HIV is elongated through the use of HAART (highly active antiretroviral therapy), which results in an increased risk of contracting cervical cancer (Health Systems Trust, 2009:1).

The relationship between HIV and cervical cancer has been established through several studies conducted in Africa on HIV negative/ HIV positive women. HIV positive women with cervical pathologies quickly progress to invasive cancer compared to HIV negative women (Health Systems Trust, 2009:1, Chirenje et al., 2002:766, Ruche et al., 1998:2401, Franceschi & Jaffe, 2007:510). Franceschi and Jaffe (2007) further emphasize the need for offering cervical screening to all women who are HIV positive so as not to miss the opportunity of preventing invasive cancer. HPV has been identified as an important risk factor for cervical cancer. Alternative methods to reduce the incidence of invasive cervical cancer resulting from HPV infection have been researched and explored. As a result two vaccines have been developed to curb the risk of cervical cancer caused by HPV. The challenge would be dissemination of information to the communities about the vaccine coupled with availability issues in resource-poor settings (Moodley, 2009:12).

2.3 CERVICAL CANCER PREVALENCE AND BURDEN

The information below provides a comprehensive overview of the global cervical cancer prevalence and burden.

(22)

The global prevalence of cervical cancer is estimated to be 2.3 million and the incidence to be 500000. Cervical cancer is the second leading cause of cancer in females in developing countries and first in developed countries. The Human Papilloma Virus (HPV) contributes to over 70% of all cervical cancer cases, of which 41-67% is high-grade lesions and 16-32% are low-grade cervical lesions. Cervical cancer screening is important to diagnose pre-cancer lesions early. The crude incidence rate and standardized incidence ratio is estimated to be 16 and 100 per 100000 women respectively. The mortality rate of cervical cancer is estimated to be 8.9 per 100000 females globally and the cumulative risk to be 0.7%. The risk of cervical cancer tends to increase with age, reaching the peak point in 35 per 100000 women from 55-64 years age groups. This rate is doubled in South Africa (WHO/ICO Information Centre on HPV and Cervical Cancer, 2007:5-14).

2.3.1 The prevalence and burden of disease in South Africa

In South Africa cervical cancer incidence has doubled compared to the global estimates and has the highest rate compared to other cancers at 30.2 per 100000 females followed by breast cancer at 27.0 per 100000 females. The number of new cases is higher in women aged 15-44 and gradually decreases with age, conversely the death rate increases with age, since pre-cancer lesions may take years to develop to invasive pre-cancer. More than 80% of cases diagnosed with HPV are found to have higher-grade lesions or cervical cancer; 88.4 and 93.8% respectively (WHO/ICO Information Centre on HPV and Cervical Cancer, 2007: 8-15).

2.3.2 The Burden of disease in KwaZulu-Natal

In KwaZulu-Natal between 1997 and 2001 about 48% of the total deaths reported were due to degenerative diseases, including the central venous system (CVS) and neoplasma. In females neoplasma was ranked the second cause of mortality contributing to about 14% of deaths. Among those females dying from cancer the most common cancer is that of the reproductive system, cancer of the cervix (KwaZulu-Natal Epidemiology Bulletin, Department of Health, 2003:4).

2.4 CERVICAL SCREENING IN SOUTH AFRICA

When the Papanicolau smear (Pap) test was introduced as one of the cervical cancer screening methods, the health authorities in most countries adopted this screening method in their annual screening policies. In developed countries this was mainly the responsibility of the

(23)

health care workers to keep a national register and mail invitation letters to women as reminders for annual smear tests when due. Historically, in developing countries, however, funds were mostly channelled to the management of the late stage cancer as opposed to early detection measures. Women were thus left at risk due to inadequate preventive care (Lovejoy, 1996: 128).

The World Health Organization has stipulated the target of women for cervical screening to be those aged 30 and above and also recommended reducing the number of cervical cancer smears to one per woman per lifetime. The South African National Department of Health proposed a total of 3 cervical cancer smears per lifetime beginning at age 30 and above and also that these are spaced at 10 year intervals. The premise behind this policy is that cervical cancer develops over time; it may take 10-20 years for a pre-cancer lesion to progress to cervical cancer. Younger women under 30 years usually present with low grade lesions that regress to normal over time. The mean age of patients with high-grade lesions is 30 years and the time estimated for progression to invasive cancer is 10 years. If a cervical cancer smear is sensitive enough it should be able to diagnose cervical abnormality early enough so as to facilitate preventive measures. Women with abnormal cervical smear results require referral for colposcopy and for further management (Department of Health: National Guidelines on cervical cancer, 2000:3).

The South African National Cancer Control Programme was adopted as the official government policy in 1997 and its implementation is decentralized to provincial and local structures. The challenge of inadequate resources e.g. poor transport systems, lack of accessible laboratories and also low uptake of this service is still evident in some provinces. The Department of Health has estimated that in order to achieve 70% coverage of the population, over five million women would need to be screened over the next 10 years. Some of the cervical screening challenges are that most women in developing countries only seek health care when they are in the advanced cervical cancer symptomatic stages (Moodley, 2009:11-12).

(24)

2.5 LITERATURE TO EVALUATE WOMEN’S KNOWLEDGE, AWARENESS AND EXPERIENCES REGARDING CERVICAL CANCER AND CERVICAL CANCER SCREENING.

Various studies have been conducted globally to evaluate women’s awareness of cervical cancer and screening practises. In a qualitative study conducted in West Virginia to evaluate awareness and knowledge of cervical and breast cancer among women, it was found that women were more knowledgeable about breast cancer than cervical cancer. Some of the common misconceptions were on the frequent use of tampons during menstruation as the cause of cancer. Some also believed that women who had a hysterectomy performed, regardless of the indication, do not require cervical screening in the future (Lyttle & Stadelman, 2006:4). These misconceptions could negatively impact on the future uptake of cervical screening services. Barriers to screening were mostly cost-related as women in developed countries are expected to pay in order to access such services. The other barriers were fear of the unknown, embarrassment, denial, lack of time and lack of transportation. Lack of motivational factors were cited and therefore there is a need for further interventions such as increased educational campaigns by health care professionals, the buddy system where the relative or friend could accompany each other to the health care service and motivation by health care providers to encourage women to go for testing. Women also advised that educational campaigns should be varied e.g. that television programs, the radio, mail and contact sessions be utilised and they also emphasized the importance of varying educational techniques by various educators to cater for different age groups (Lyttle & Stadelman, 2006:4-5).

In a quantitative study conducted to assess knowledge, perceptions and attitudes of health care professionals in the U.S.A, findings were that most professionals are knowledgeable about risk factors of cervical cancer. However, they cited the lack of time and resources in offering educational information on cancer to patients. Preventive counselling to reduce the risk of cervical cancer was viewed as of less importance to the professionals (Tessarro, Herman, Shaw and Giese, 1996:272). This implies that no matter how knowledgeable health care professionals are, sharing information with patients and communities is of greater benefit to reduce the risk of cervical cancer.

(25)

Most studies conducted globally have indicated lack of knowledge and awareness in women regarding cervical smear testing and cervical cancer. This has resulted in a low uptake of screening services. In a qualitative study conducted among women aged between 21 and 56, most women had a poor knowledge of cervical cancer (Wong, Wong, Low, Khoo and Shuib, 2009: 51). Most of them had heard of cervical cancer previously but were not aware that it could be prevented. Only a few older, married women were aware that if cervical cancer is detected early it could be managed. The role of the Pap smear in early detection was also poorly understood. Only less than half of women recognized the risk factors. Most surprisingly, most of these women thought the Pap smear is performed to diagnose sexually transmitted infections, and the procedure itself was not well understood (Wong et al., 2009: 51). These findings indicate that a lot needs to be done to increase public awareness, knowledge and insight of women regarding the importance of early screening for cervical cancer. Educational messages should reach all women regardless of educational level, race or socio-economic status. Most of the women who participated in the above study had a secondary education. This shows that the educational level would not contribute much to the assimilation of health education messages.

In a study conducted in Nigeria amongst professionals, other public servants and students, the level of awareness of cervical screening was average (52.8%), however only 7.1% had ever done a Pap smear test. The most common reasons given for not doing a Pap smear test included the lack of awareness (46%), fear of a bad result and some felt there was no need for it (Ezem, 2007:94-96). One would expect the level of awareness to be higher in this sample. However, this result attests to findings from other countries. Women do not access screening services despite awareness about such services. In most cases the proportion of women who had ever had a Pap smear is far less than half of the women who reported an awareness of the Pap smear. Almost half of the women in a study conducted in Ghana reported doing a Pap smear, because the doctor asked for it as part of patient management. Results also showed that although 93% of women have heard of cervical cancer, only 37% of those had adequate knowledge about this disease. Of those with adequate knowledge, only 8.5% have ever had a Papanicolau (Pap) smear done (Adanu, 2002: 487). In a study to evaluate availability of screening services in African countries it was found that 95% of health institutions at all health care levels had basic infrastructures to perform cervical cancer screening (Chirenje, Rusakaniso, Nyamapfeni, Kamba and Padian, 2001:127-128).

(26)

Earlier studies conducted in South Africa (Cape Town) also revealed a lack of knowledge on cervical cancer and Pap smears. None of the women interviewed knew what a cervix was; however, they could easily recognize the word Pap smear and link it to cancer of the uterus. Most women thought the Pap smear was done to diagnose infections, infertility and also to clean the uterus following severe vaginal bleeding or miscarriage. Only a few participants linked Pap smears to cervical cancer diagnosis. The reasons for doing a Pap smear are mostly linked to ill health and also as per nurse instructions. This indicates that the use of the Pap smear as a screening test for early pre-cancerous lesions is poorly understood. Some of the obstacles to Pap smear verbalized by women are: fear, embarrassment, lack of time and stigma (Abrahams, Wood and Jewkes, 1996:14-21). The study conducted in the rural KwaZulu-Natal population showed similar results (Hoque, Hoque and Kader, 2008:113-115). The uptake of cervical screening is still very low despite these services being cost-free in most regions. Research evidence suggests that factors such as knowledge, attitude and awareness play a vital role in the utilizing of this service. There is also no significant difference between women who attend and non-attenders in respect of knowledge (Twinn, Shiu and Holroyd, 2002:379). This indicates that lack of education and knowledge does not exist as a main barrier to women attending clinics for cervical cancer screening. In a survey conducted to assess barriers and women experiences to pap smear testing in the same geographic area in KwaZulu-Natal, women displayed that they had information about the need for cervical screening. In almost all women interviewed, the motivator for cervical screening was the invitation they got from the professionals and also the pelvic symptoms they experienced (Wood & Jewkes, 1996:9). The same study found an association between promiscuity and cervical cancer, which may result, in low utilization of the screening service (Wood & Jewkes, 1996:10). A recent study (2008) conducted in one of the tertiary institutions in KwaZulu-Natal reported a lack of knowledge regarding cervical cancer and cervical screening. Less than half of the students had heard of cervical cancer and the risk factors were poorly understood. Half of the students mentioned HPV as the main cause of cervical cancer. The main reasons for not doing a cervical screening test were fear and being healthy (Hoque & Hoque, 2009: 21-24). The above implies that there is a need for health education and making the communities aware of the importance of cervical cancer screening so as to increase the utilization of cervical screening services. Unlike the other countries, the issue of cost as a barrier to screening was perceived differently in KwaZulu-Natal; professionals who interviewed patients did not identify the cost for the smear as a problem. They explained that

(27)

women often interpret “no charge” with poor quality service. This is unfortunate as most women benefit from free primary health services and the bulk of the women in this area are unemployed. Barriers that surfaced were mostly the lack of trained staff to provide service and education, as well as women’s reluctance and fear caused by misunderstanding and lack of motivation and anxiety. The other important barrier was the lack of adequate communication with clinicians, especially of the referring hospital. This then had a negative impact on an appropriate follow-up care of patients. In addition, when referred, women often resisted attending further investigations following abnormal smears (Wood & Jewkes, 1996: 14-15).

2.6 ACCESS AND UTILISATION OF CERVICAL SCREENING SERVICES

The prevalence of cervical cancer has been greatly reduced through successful cervical screening programmes especially in developed countries. Most developing countries are still faced with challenges of implementing such services. Success depends on a number of factors, such as coverage of the right target women, excellent follow-up service, early management of pre-cancer lesions and resource availability (Pollack, Balkin, Edouard, Cutts and Broutet, 2007:57). Research conducted in other countries among women who had cervical cancer has shown a significant relationship between age and failure to screen, older women are more likely not to screen compared to younger women. The other important factors to be considered to ensure that cervical screening contributes to prevention of invasive cancer are: diagnostic failure whereby the screening test is unable to detect precancerous lesions due to poor techniques. The aim also is reaching a greater percentage of women who have not had a recent Pap smear. This is to diagnose undetected precancerous lesions early before the condition develops into cervical cancer (Leyden, Manos, Geiger, Weinmann, Mouchawar, Bischoff, Yood, Gilbert and Taplin, 2005:677). There is a need to periodically review health services to ensure that all factors influencing access to these services are addressed.

Studies conducted in Africa show that above 95% of health care institutions have basic infrastructures to conduct cervical screening. Most screening was provided on demand through family planning services. The most common barriers to providing cervical screening tests were a lack of a clear policy and procedures, and also staff competencies. Most smears were evaluated in tertiary hospitals and this caused delays in turnaround times of results.

(28)

Most women commonly screened were below 25 years and also women with invasive cervical cancer (Chirenje et al., 2001:128-129).

Data from the Western Cape, South Africa, indicates that in the year 2005, 45 997 (74%) of the total women screened were over the age of 30 years. Abnormalities were detected in 7.3% of the Pap smears taken. This is encouraging although much needs to be done to reach 100% of eligible women. In KwaZulu-Natal 28 760 Pap smears were performed in 2005, accounting for 26% of the target. This indicates that a large pool of women remains unscreened. Reaching the 70% National target coverage remains a challenge. The other factor is that a low percentage of women who have ever done a smear test are not informed of the purpose of the test (Moodley, 2009:12).

2.7 SUMMARY

From this comprehensive review of literature it is apparent that cervical cancer screening knowledge is essential as a health promotion strategy. The burden of disease due to cervical cancer is high and also poses a challenge to the overburdened health care systems in developing countries. This review formed the basis for the questionnaire and the research design that was utilized in this study. Few studies have been conducted in this province to assess the level of knowledge, awareness and attitudes of women to Pap smear testing and cervical cancer. This study will add to the body of knowledge, which would help in re-enforcing prevention messages and cervical cancer information. It would also provide a basis for exploration of key issues through a qualitative approach in the future. In the next chapter the methodology implemented for this research study which explored women’s awareness, knowledge and experiences regarding cervical cancer in rural KwaZulu-Natal, South Africa were discussed.

(29)

CHAPTER 3:

METHODOLOGY

Chapter two provided a comprehensive overview of the literature on the topic: women’s awareness, knowledge and experiences regarding cervical cancer and screening and related matters. This chapter describes the methodology that was used in this study. According to Burns and Grove (2001:26) research methodology is “the application of all steps, strategies and procedures for gathering and analysing data in a research investigation in a logical and systematic way”. This can be defined as the research plan that guides the research. It involves a research design, sampling methods, data collection methods and analysis, and ethical considerations.

3.2 THE PURPOSE OF THE STUDY

The aim of this study was to assess women’s awareness, attitudes and experiences regarding cervical cancer smear screening and cervical cancer in rural KwaZulu-Natal, and to better understand the factors that influence access to and utilization of cervical screening services by rural women. In resource-poor settings, access to services offering cervical screening is still a challenge and according to Adanu (2002:486), it is estimated that more than 50% of women in developing countries have never had a single screening test for cervical abnormalities.

3.3 THE RESEARCH OBJECTIVES

This study was guided by the following objectives to:

1. determine the rate of screening for cervical cancer among women aged 30 years and above in rural KwaZulu-Natal;

2. explore and describe the knowledge and perceptions of women on cervical cancer and cervical cancer screening in this rural population.

3. identify and characterize socio-demographic factors associated with accessing screening services in women aged 30 years and above and

(30)

3.4 THE RESEARCH DESIGN

A research design “guides the researcher in planning and implementing the study in a way that is most likely to achieve the intended goal” (Burns & Grove 2001:223). This was a descriptive study, using a questionnaire to generate quantitative data to gain insight into participants’ awareness, knowledge and experience relating to cervical cancer and cervical cancer screening.

A descriptive study aims to quantify the problem, giving detailed information and also taps into perceptions of communities and groups. Some literature describes descriptive studies as hypothesis-generating studies since they bring out questions for further research about situations. The benefit of conducting such studies is that information obtained assists direct planners and service providers in service planning and resource allocation (Katzenellenbogen, Joubert & Abdool-Karim, 1999:89).

This was a descriptive study using a structured questionnaire (Appendix A) to collect data and some open-ended questions which were added to the questionnaire to further explore opinions and perceptions. Experienced and knowledgeable registered nurses in the Microbicide clinical trial clinics who were utilized for research and the researcher administered the survey questionnaires. These nurses were already trained in cervical cancer screening since it was offered as a standard of care to all Microbicide clinical trial participants. In addition the researcher conducted extra two days training on the cervical cancer protocol including the questionnaire and the informed consent process. Training sessions were evaluated through mock interviews, where data collectors were asked to conduct interviews on each other using the study questionnaire. This helped the researcher to identify any training gaps that had to be addressed before actual data collection, e.g. correct adherence to interviewer instructions, skip questions, correct phrasing of questions and recording.

3.5 STUDY POPULATION AND SAMPLING

The Africa Centre for Health and Population studies is one of the 6 sites involved in a multicentre trial-Double-blind, placebo-controlled Microbicide clinical trial, testing Microbicide gel –PRO-2000/5 for safety and efficacy in preventing vaginally acquired HIV. This site is situated in UMkhanyakude District, Mtubatuba, in rural KwaZulu-Natal. The Microbicide trial commenced in March 2006 with a target to enrol 1200 women and ended follow-up in August 2009. Women were recruited from district family planning and

(31)

immunisation clinics. Those who met the inclusion criteria were randomised into three study arms, namely PRO 2000 0.5%, PRO 2000, 2% and Placebo gel. Women enrolled in this trial were followed up over a period of a year, returning to clinic every month for follow-up care. They completed follow-up at different time points. The last group of women completed follow-up on 28 August 2009. All the women who participated in the Microbicide clinical trial were offered cervical cancer screening through a Pap smear at enrolment as a standard of care procedure in this trial. The researcher then identified this group of women as a sample for the cervical cancer study, to further explore the role of prior screening and knowledge on this subject.

The initial target for the cervical cancer study was 120 participants, 10% of the Microbicide trial participants. However there were delays in Ethical committee approval, the cervical cancer protocol was initially submitted in June 2008 and only granted Ethics approval in June 2009. Data collection commenced on 06 July 2009. Only six hundred women were still on clinical follow-up during that period, a convenience sample of 69 women who agreed to be part of the cervical cancer study was obtained. Women were recruited from the three clinics at the UMkhanyakude district in rural KwaZulu-Natal, Madwaleni, Mtuba and KwaMsane clinics. This was a convenience sample as eligible women who were already randomized to the Microbicide clinical trial were asked to participate as they came to the clinics.

3.5.1 Inclusion Criteria

The following inclusion criterion was used to enrol women to this study:

 Women enrolled in the Microbicide clinical trial attending clinic for Microbicide trial follow-up care between 06 July and 07 August 2009.

 Participants attending MDP trial clinics aged 30 years and above.

 The three clinics selected for sampling were all in the UMkhanyakude district in rural Kwazulu-Natal province.

3.6 RELIABILITY AND VALIDITY

Assessment of reliability is done to ensure that the research instruments measure consistently and accurately with repeated measures (Gerries & Lacey, 2006:376). Katzenellenbogen, Joubert & Abdool-Karim, 1999, defines reliability as “the degree of similarity of the information obtained when a measure is repeated”. The researcher tested the questionnaire by giving it to colleagues who are working in the same clinics (these colleagues were not part of

(32)

the study) to determine if all interpreted the questions similarly. According to Katzenellenbogen, Joubert and Abdool-Karim 1999:124), validity is defined as the ability of an instrument to measure the variables that it is intended to measure. The questionnaire was piloted on a sample of 5 women, aged 30 years and above, attending the family planning (reproductive health clinic) at Somkhele clinic. Assessment of reliability is done to ensure that the research instruments measure consistently and accurately with repeated measures (Gerries & Lacey 2006:376). Katzenellenbogen, Joubert and Abdool-Karim (1999:124), define validity as the extent to which a measure actually measures what it is meant to measure. Attempts were made to improve and evaluate validity and reliability in this study. This was ensured through standardization of the method of data collection, training of research nurses who were collecting data, supervision and control.

Nurses who were assisting with data collection were trained on the cervical cancer protocol to ensure the high quality of data and also standardization of data collection. Quality control of completed questionnaires was done by the researcher. The researcher checked questionnaires for completeness, consistency and legibility and where necessary meetings were held with the interviewers to clarify inconsistencies. Inconsistencies identified and clarified at the beginning of the study were partial adherence to interviewer instructions, e.g. incorrectly asking skip questions.

3.7 THE RESEARCH INSTRUMENT AND DATA COLLECTION

Data collection was through a self-designed structured questionnaire. The questionnaires were formulated in English and translated into the language that is spoken by the target sample i.e. isiZulu by the research nurses who are trained registered nurses and the researcher during an interview (Cervical Cancer Questionnaire: Appendix B). The instrument was a self-designed structured questionnaire with mainly closed-ended questions and a few open-ended questions. The questionnaire was typed legibly and was divided into 3 sections, namely section 1: Socio-demographics, section 2: knowledge/awareness, section 3: perceptions and experiences. Socio-demographic characteristics considered were age, marital status, educational status- highest level attained, employment status, earnings, and sexual/reproductive history. Knowledge and awareness was obtained from general knowledge questions on cervical cancer description, whether it is preventable and how, the cervical screening test type and frequency of cervical cancer screening tests. Perceptions and

(33)

experiences were elicited from the history of previous cervical screening, experiences and willingness to screen for cervical cancer in the future.

Research nurses who assisted in data collection are registered nurses who received two days additional training on cervical cancer protocol, the informed consent process and questionnaire administration. The extra training was conducted to ensure good data quality and standardization of the data collection process. Staff training was evaluated through simulated interviews. Literate participants were allowed to complete a questionnaire on their own with the assistance from the interviewer. The interviewer and the literate woman who chose to record her own responses both had copies of the questionnaire. The interviewer then read out questions systematically to ensure the woman answered all questions as required. Women were asked (open and closed questions) about their knowledge, perceptions of cervical cancer and cervical screening. The history and experiences of previous cervical screening were sought as well as barriers to cervical screening were identified.

3.8 THE PILOT STUDY

A pilot study is defined by Van Ort (1981), quoted by Burns and Grove (2009:44), as a “smaller version of a proposed study that is conducted to refine the methodology”. It could be performed in various stages of a research process. In this study this only involved piloting the questionnaire during the early stages of the questionnaire formulation.

The research instrument was piloted on a sample of five women who were comparable to sample respondents in terms of socio-demographic and geographic characteristics. This was conducted in July 2008 at Somkhele Clinic and resulted in the following benefits: the time taken to complete the questionnaire was established; statements and questions that were misinterpreted by the respondents were re-worded and corrected.

3.9 THE ROLE OF THE RESEARCHER

The researcher was involved in protocol development, questionnaire design, the training of three registered nurses from each clinic on the cervical cancer protocol, a standardized method of data collection including administering the consent, data collection. The researcher also did quality control of data collection, analysis of data and report writing. All completed questionnaires were sent to the researcher at the end of each day to be quality controlled for completeness, accuracy and consistency. Data was then captured on an excel spread sheet by

(34)

the researcher in preparation for data analysis. Data analysis was performed once data capturing of all questionnaires were complete.

3.10 ETHICAL CONSIDERATIONS

Research involving human subjects should always be guided by good clinical practice and human right principles to ensure protection of participants. Some of the ethical responsibilities of researchers are to maintain privacy, informed consent-ensuring that there is voluntary participation, protection of participants, informing participants what the study is for, how information will be used and whether there is any potential risk expected (Katzenellenbogen et al., 1999:89).

This study which explores women’s awareness, knowledge and experiences with regards to cervical cancer in rural KwaZulu-Natal was approved by the Stellenbosch University, Faculty of Health Sciences, Human Research Ethical Committee (HREC) on 9 March 2009 and 2 April 2010 respectively (Appendices C01 and C02) and the study site project leader. The respondents were asked to be part of the study after detailed information pertaining to the study was explained to them. They were asked to sign an informed consent form as proof of their voluntary participation.

3.10.1 Confidentiality and Anonymity

The researcher used screening and trial numbers already allocated to these participants as identity numbers, since no names were used on any of the questionnaires except informed consent documents. Full participant names were used on the informed consent forms, and these were kept locked separately from the questionnaires. The participants were assured of confidentiality of their information during an informed consent process and privacy was maintained throughout. Completed questionnaires were transported from the clinics by the Africa Centre driver to the office in a well secured box to avoid unnecessary loss. Data capturing was done by the researcher and the electronic spread sheet was saved in a password controlled excel folder.

3.10.2 The consent

The study information sheet was translated into the language spoken by participants, namely, isiZulu, and was made available to all participants (IsiZulu participant information sheet and consent: Appendix C03. Illiterate participants were assisted by interviewers to read the information sheet to them. Participants were encouraged to ask questions to ensure clarity of

(35)

the study information. At the end of the information session volunteers were asked to sign the consent document as proof of voluntary participation and also to acknowledge their understanding of the study information. Illiterate participants were asked to put a thumb print as a signature. The information sheets and a copy of the signed informed consent form were provided to participants to take with them.

3.11 DISSEMINATION OF RESULTS

The researcher has an ethical obligation to make research results available to all stakeholders; this includes participants, district clinic staff, the district hospital and the local community. The researcher will use feed-back systems in place at the Africa centre for this purpose such as community meetings, formal meetings with the hospital board and clinic staff, road shows and the Umbiko community magazine. A copy of the final report will be made available to the Hlabisa district hospital.

3.12 SUMMARIZING THE SAMPLE DATA

The primary outcome measure for this analysis was whether these women had ever been screened for cervical cancer (Q13, section 3-have you ever screened for cervical cancer (prior to joining the trial?) The primary outcome measure was assessed from the history of previous cervical screening obtained, which was either “yes” or “no”. The secondary outcome measure employed in the analyses reflected knowledge and perceptions of women with regards to cervical cancer and cervical screening. Knowledge was determined through the general knowledge about cervical cancer, cervical cancer risks and cervical screening, Q7-12: Section 2: Q7. What do you understand by cervical cancer? This was an open-ended question where respondents were expected to state in their own words what cervical cancer is., Q8 what basic test is conducted to screen for cervical cancer? Respondents were expected to choose from 3 responses given. Q9 Do you have information about cervical cancer screening, if so where did you get this information? Q10 Mention at least 3 risk factors for cervical cancer, Q11 Can cervical cancer be prevented? “Yes” “no” or don’t know. Q11a if yes to q11, how? Respondents who knew cervical cancer could be prevented were expected to state how in Q11a. Q12-How often do you think women should screen for cervical cancer- four responses were given to choose from. If a woman had reported previous screening her experiences and also intention to screen in the future were explored, also if no prior cervical screening was reported reasons were explored (Q13-15b). Q13a if answer to Q13 (ever screened for cervical cancer) was “no” women were expected to give reasons-choose from responses given. Q13b

(36)

if answer to Q13 was “yes” the woman was expected to share her experiences, through choosing from responses given or specify. Q14 cervical cancer risk established-“Do you think you are at risk of cervical cancer? Q15-15b were on establishing future intentions to screen- Q15 Would you screen for cervical cancer in the future?, and also where the woman intended to go for screening Q15a, if not reasons were established, Q15b. Socio-demographic factors associated with previous screening for cervical cancer were explored to establish associations. Demographic and socio-economic characteristics included age, highest educational level achieved employment status, monthly earnings and marital status. Age was assessed as categorical variable (30-34, 35-44, 45+, year age groups) throughout the analysis. Highest educational level achieved was also assessed as categorical variable (no education, primary education, secondary education, matriculation or above). In addition, education was evaluated as a categorical variable (no education, primary, secondary or higher throughout). Employment status was categorized as a binary variable namely being unemployed versus being in either part-time or full-time employment. Earnings (women who were employed) were classified as categorical variable (R500 or less, R501-R1000, R1001-R2000, R3000+). Marital status was treated as a binary variable of being married or unmarried.

Other variables included age of sexual debut, parity and contraception. The age of sexual debut was represented by categorical variable with sexual initiation below 15, between 16 to 19, or 20 + years of age. Parity was reported as ordinal categorical variable (0, 1-3, 4-5, 5+). Contraception usage was reported as binary variable of using or not using any method of contraception. Contraceptive type was also reported as (pills, injectable, sterilization, condom or other).

All questionnaires were inspected for completeness and consistency of responses e.g. whether all questions were answered, skip questions followed accurately, legibility of handwriting. Questionnaires were transported by the Africa Centre driver from the clinics to the Africa Centre (where the researcher works) each day and captured into the Microsoft Office Excel 2007 spread sheet by the researcher.

3.13 DATA ANALYSIS

Data was extracted from Microsoft Excel 2007 and analysed in STATA 10 (Stata Corporation, College Station, Texas, USA) and the SAS statistical programme for Fishers Exact test. The above test is used to classify objects in two different ways (categorical data) to

(37)

examine the significance of the contingency between the two kinds of classification (Fisher, 1922:87-94). Descriptive statistics (mean, standard deviation, percentages and frequency distribution) was used to analyse quantitative data to determine central tendency and variation in the data. The overall proportion of women screened for cervical cancer through a Pap smear was established. In addition this proportion was broken down per socio-economic, demographic and sexual behaviour variables.

The relationship of periodic screening for cervical cancer with demographic, socio-economic characteristics was evaluated in bivariate analysis using the Pearson-Chi Square test. Statistical significance between the group who had been screened and those who have never been screened for cervical cancer was analysed using Fisher’s Exact test and expressed in odds ratio, p-value and the confidence interval. The p-value was set at a significant level <0.05. For all values less than 5, the Chi square test was not sensitive to pick up differences, thus the Fisher’s Exact test was used to determine p-values (age-group, educational level, sexual debut, contraceptive method and pregnancies).

3.14 LIMITATIONS OF THE STUDY

This study enrolled women who were recruited from the district family planning and immunisation clinics and participating in the Microbicide clinical trial. These women had to fulfil a certain criteria to be randomised to the Microbicide clinical trial. Some of the basic enrolment criteria used was: They had to be HIV negative, non-pregnant, sexually active (at least 2 sex acts in a week), willing to receive condom use counselling, age 18 years and above, willing to attend clinic for scheduled follow-up visits, etc. Information and selection bias might have occurred as women were recruited from primary health care clinics, so this study might have underestimated patients who never sought care and those who normally seek care from other providers. The researcher initially planned to recruit and enrol 120 participants. However, due to time constraints, the final sample size was 69 participants. Logistical issues of limited time negatively influenced the sample size. The small sample size may have masked real differences between the groups that could not be detected due to very wide confidence intervals in some analyses. As a result of the small sample size the results could not be generalised to the population.

(38)

3.16 SUMMARY

In this chapter a comprehensive description of the research design, sampling procedure, data collection technique, data management and analysis methods have been explicitly presented. Data obtained from these analyses will be presented in the next chapter.

Awareness, knowledge and experiences of women regarding cervical cancer in rural KwaZulu-Natal, South Africa